Reference no: EM133784180

Forecasting and Quantitative Analysis

Assessment:

Question 1
The below output shows the results of an ARIMA model which was fitted to a time-series data of Australia's GDP, and an excerpt of the data. Based on the below model and information:

## Series: GDP
## Model: ARIMA(2,1,0)
## Coefficients:
## ar1 ar2 constant
## 1.50.2 0.6
a) Explain what the time series is likely to look like (i.e., cyclical, seasonal, with a trend).
b) Calculate the predicted GDP value for Q4, 2023?

Question 2
Two models, A and B, are evaluated on the basis of the residual diagnostics in Figures A and B. Explain which model performs better and why, by discussing: the residual plot, the ACF, and the residuals' histogram.

Question 3
The following monthly sales of coffee (in thousands of AUS dollars) have been recorded for March, April, May and June. Examining the forecasting accuracy for the month of June only, explain which of the following forecasting method would you recommend: the Naïve method, the Average method, or the Simple exponential smoothing method (assuming alpha=0.6 and initial state of 58)?

Make sure you load the fpp2 and fpp3 library before you answer Questions 4-6.

Question 4
This question requires you to use the dataset, global_economy. Your task is to analyse the Imports for United Kingdom and use at least two forecasting techniques you have learnt in the course to forecast Imports for the next 10 years. Pay attention to whether the data needs to be transformed. Perform model evaluation. Discuss all your results carefully.The analysis needs to be thorough.

Question 5
This question requires you to use the dataset, aus_production. Your task is to 1) conduct a thorough analysis ofCement production; 2) use an appropriate ARIMA (from the class of ARIMA models you have learnt in the course) to forecast Cement production for the next 4 years. Pay attention to whether the data needs to be transformed. Perform model evaluation Discuss your results carefully. The analysis needs to be thorough.

Question 6
This question requires you to use the dataset, us_change that contains information on percentage changes in quarterly personal consumption expenditure, savings and a few other variables.Your task is to forecast Consumption based on Savings. You will use the dynamic regression approach (regression with additional ARIMA errors) to predict Consumption for the next 10 quarters. After estimating the regression model, when making predictions, assume that the future percentage changes in savings is 2%.Evaluate the model.